Antecedent flow conditions for MS

MS: Spatiotemporal variation in mountain stream metabolism and nitrogen cycling across contrasting flow regimes

Pause to make a column for biomass to chla ratios And also make biomass in in micro grams to more comparable to chl-a

Additive columns for combined surface and pore water nutrient as tot_ in ugL

Kernal density plots of GPP and ER

Colored by years

Modified figure 2

Focal Question:

How does stream metabolism vary based on antecedent flow conditions, and nitrogen availability across small and large stream?

Can we frame wet years as a shrinking productivity window? Or is it more a function of a mismatch in phenology ques, where peak light occurs at unsuitable streamflows (e.g. to high) for epilithic algae?

Ideas:

1. Figure out what flow threshold at each site we can’t really model GPP at?

1. How large is that window from year to year?

2. How is the cumulative GPP during that time?

3. How is the cummulative PAR and water temp?

2. Center dynamics on that flow threshold

1. X-axis could be time since peak “snow pack metric” or “flow metric”

3. Double check is there evidence of UV suppression, should double check magnitude of surface PAR

Data resolution for DO and metab

DO in black, metabolism in green

### big grid 
WY_n_grid2 <- ggarrange(
  templt,
  templt_BWU,
  templt_GBL,
  templt_GBU,
  ncol = 1, nrow = 4)

WY_n_grid2

Get snotel data

Rough time series plots to start getting at the antecedent flow conditions

TS_plot <- ggplot(covariat_datq%>%mutate(DOY=yday(date))%>%filter(site=="BWL")%>%filter(water_year<2024), aes(x = DOY, y = PAR_surface, color =as.factor(water_year), shape = site)) +
  geom_point(size = 2, alpha = 0.5, position = position_dodge(width = 0.3)) +
 # geom_line(aes(x = DOY, y=PAR_inc )) +
    geom_line(aes(x = DOY, y=(Q_m*100))) +
 scale_color_viridis_d(option = "viridis") +
  geom_point(size = 2, alpha = 0.7) +
  #scale_color_manual(values = siteC_colors) +
  scale_shape_manual(values = c(15, 0, 17, 2)) +
  ylab(expression(PAR~at~stream~surface~(mu~mol~m^-2~s^-1))) +
  theme_classic() + facet_grid(site~.)

TS_plot

TS_plot1 <- ggplot(covariat_datq%>%filter(water_year<2024), aes(x = Q_m, y = GPP_mean, color =as.factor(water_year), shape = site)) +
  geom_point(size = 2, alpha = 0.5, position = position_dodge(width = 0.3)) +
 # geom_line(aes(x = DOY, y=PAR_inc )) +
 #scale_color_viridis_d(option = "viridis") +
  geom_point(size = 2, alpha = 0.7) +
  #scale_color_manual(values = siteC_colors) +
  scale_shape_manual(values = c(15, 0, 17, 2)) +
  #ylab(expression(PAR~at~stream~surface~(mu~mol~m^-2~s^-1))) +
  theme_classic() + facet_grid(site~.)

TS_plot1

(1) Figure out what flow threshold at each site we can’t really model GPP at?

flow stats for when GPP was modeled

Max Q where GPP was obtained…

• BWL : 2.763 cms

• BWU : 5.990 cms

• GBL : 0.622 cms

• GBU : 0.299 cms

Get DOY dates for max light and flow

Manually identify baseflow onset

Looking at Cumulative GPP

Dotted lines for onset of baseflow Dashed lines for peak PAR

Looking at cumulative GPP + light + flow at each reach:

BW Lower

BW Upper

GB Lower

GB Upper

Leaf area index 2021-2023

Surface PAR 2021-2023

Peak spring Q 2021-2023

Code sandbox: